Skip to content

Latest commit

 

History

History
29 lines (24 loc) · 1.5 KB

README.md

File metadata and controls

29 lines (24 loc) · 1.5 KB
Raw

This is OpenCV implementation of integral histograms as described in [1]. It can be used for object detection and localization. We used it for change detection with LTP textural descriptors. Because of this our implementation is in some aspects specialized for this problem.

We implemented three specific ways of calculating the integral histogram:

  • integralHistogram(): Calculates the integral histogram on one input image.
  • integralHistogramVM(): Calculates the integral histogram on two images. The pixel intensity of Val image determines the bin index and the pixel intensity of Mag determines the increment in magnitude of this bin.
  • integralHistogramJoint(): Calculates the integral histogram on two images. The resulting histogram is two dimensional and is build in the sense of joint PDF.

Once your integral histogram is calculated, you can get histograms of any rectangular portion of the original image using function regionHistogram(). This histogram is calculated in linear time with respect to number of histogram bins.

Using the compare() function you can compare two images using histograms calculated in a sliding window of arbitrary size. For reasonably sized images this can be done in realtime.

See main.cpp for example usage.

[1] PORIKLI, F. Integral histogram: a fast way to extract histograms in Cartesian spaces. In: 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05). IEEE, 2005, 829-836 vol. 1. DOI: 10.1109/CVPR.2005.188. ISBN 0-7695-2372-2.